EFFECTS OF DIVERSITY ON COMMUNITY ASSEMBLY DYNAMICS IN NEWLY FORMED POND COMMUNITIES
BY
CHRISTOPHER J. HOLMES-SINGH
THESIS
Submitted in partial fulfillment of the requirements for the degree of Master of Science in Biology with a concentration in Ecology, Ethology, and Evolution
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2014
Urbana, Illinois Master’s Committee:
Professor Carla Cáceres, Chair Professor James Dalling
ABSTRACT
Theory suggests that the initial level of genetic or species diversity in a new habitat will influence community assembly dynamics. Previous studies have shown that species diverse communities can buffer against invasion by heterospecifics. In addition, the monopolization hypothesis predicts that a genetically diverse population will be more likely to adapt to the local environment, monopolize resources, and buffer against
invading competitors. This rapid evolution of a focal population can enhance priority effects thereby decreasing invasibility of later arriving colonists (termed evolution-mediated priority effects). Nevertheless, empirical investigations of these diversity effects, at both the genetic and species level remain rare, especially for animal systems.
To test this theory, we conducted a field experiment in which initial stocking diversity (both intra- and interspecies) of freshwater zooplankton in newly constructed pools was manipulated in a 2x2 fully factorial design. Zooplankton communities and several abiotic variables were sampled every two weeks (from May to August) for 3 years. Estimates of overland dispersal were measured in the second year of study. We also conducted laboratory assays on ecologically relevant traits for our focal Daphnia pulex clones originally stocked in the field experiment to determine if performance in the field (ability to persist) was determined by genotype differences and/or phenotypic plasticity in the population growth rate parameter, r. Despite theoretical predictions, after 3 years we found no difference in taxonomic richness or diversity among stocking treatments. A total of 31 species were recorded in the metacommunity with an average cumulative taxonomic richness ranging from 6.1 to 7.6 species per pool. Dispersal of zooplankton taxa was rapid with 8 taxa dispersing in 7 days, but we found no difference
in the number of dispersing propagules based on number of neighboring source pools. Pools in the same patch, delimited into landscape “hexagons”, had similar community composition and variation in community structure was explained by average pH and dissolved oxygen concentration. Populations of our focal species, Daphnia pulex, exhibited rapid evolutionary change after the second year of study. However, performance in the field was not explained by genotype differences in intrinsic r
measured in the laboratory assays on high food, but may be explained by varying levels of plasticity in r among clones. Our study demonstrates that despite theoretical
predictions that genetic diversity of a focal population and initial species diversity should influence community assembly dynamics, they do not appear to have an effect on early successional community species richness or diversity, or population persistence of the focal species D. pulex. Local abiotic factors or differences in dispersal rates among taxa may have a larger influence on community assembly dynamics in freshwater zooplankton metacommunities.
Dedicated to my loving grandparents Drs. Kenneth and Linda Holmes PhD, parents Dr. Nirmal Singh and Kari Singh,
ACKNOWLEDGEMENTS
I would like to thank my advisor, Dr. Carla Cáceres for her continued
encouragement, guidance, and expertise. My committee members, Dr. James Dalling and Dr. Kimberly Schulz provided immeasurable expert advice and support for this project and for that I am also very thankful. I thank Dr. Carol Augspurger for her invaluable guidance while I was an undergraduate at the University of Illinois at
Urbana-Champaign. This project would not have been possible without laboratory and field assistance provided most graciously by John Crawford, Kelly Hogan, Hannah Wright, Liliana Calderon, Glynn Davis, Frankie Rodriguez, James Arrigoni, Tara Stewart, Ping Lee, Ilona Menel, and Jessica Kirkpatrick. I would like to further thank Jessica
Kirkpatrick for her providing comments on drafts as well as her loving support and encouragement. Last but not least, I would like to thank Dr. Ken Paige, Lisa Smith, and the Department of Animal Biology for providing me this opportunity to enter this program as a Master’s student and their continued support throughout this experience. Grants from Sigma XI (GIAR), Department of Animal Biology, School of Integrative Biology, University of Illinois Travel Grants, Odum-Kendeigh Research Award, and NSF EAGER grants DEB-0947314 & 0947245 made these research projects possible.
TABLE OF CONTENTS
CHAPTER 1: EFFECTS OF INITIAL STOCKING DIVERSITY ON COLONIZATION
DYNAMICS... 1
1.1 INTRODUCTION ... 1
1.2 MATERIALS AND METHODS... 5
1.3 RESULTS... 9
1.4 DISCUSSION ... 12
1.5 TABLES AND FIGURES ... 18
CHAPTER 2: GENETIC DIFFERENTIATION OF DAPHNIA PULEX IN FRESHWATER PONDS: RAPID EVOLUTIONARY CHANGE AFTER TWO YEARS... 24
2.1 INTRODUCTION ... 24
2.2 MATERIALS AND METHODS ... 27
2.3 RESULTS ... 31
2.4 DISCUSSION ... 33
2.5 TABLES AND FIGURES ... 39
REFERENCES... 42
APPENDIX A: OTHER FIGURES FOR CHAPTER 1 ... 55
APPENDIX B: STATISTICS FOR CHAPTER 1 ... 58
CHAPTER 1: EFFECTS OF INITIAL STOCKING DIVERSITY ON COLONIZATION DYNAMICS
1.1Introduction
Community structure results from the interaction of historical (priority effects) and on-going processes operating at both local and regional scales. Given this
complexity, generating accurate predictions for the outcome of assembly remains
difficult for most systems (MacArthur and Wilson 1967, Gadgil 1971, Young et al. 2001, Weiher and Keddy 2001, Rajaniemi et al. 2012, Dickie et al. 2012). The importance of both abiotic and biotic factors on establishment success is well known, yet the relative importance of these multiple local processes varies significantly among systems (Shurin 2000, Cottenie and De Meester 2004, Vanormelingen et al. 2008, Algar et al. 2011). Further, there are two schools of thought regarding the importance of history (timing and order of colonization) in determining final community structure. Some models and experiments (e.g., Neill 1975, Tilman et al. 1986, Sommer 1991, Law and Morton 1996) suggest that community structure should converge on a similar state when habitats are similar, irrespective of colonization order. In contrast, multiple states equilibrium may be a product of colonization history (Robinson and Dickerson 1987, Drake 1991, Luh and Pimm 1993, Samuels and Drake 1997, Law 1999, Chase 2003). Understanding the relative importance of these factors underlying community assembly is necessary to accurately predict if restored or newly created habitats will achieve their intended community form and function (Loreau et al. 2001, Chase 2010).
Community ecologists have long considered the importance of interspecies interactions on the abundance and distribution of species over both time and space
(Hutchinson 1959, 1961, Connell 1961, Paine 1966, MacArthur 1967, Sih et al. 1985, Drake 1991, Case 1991, Tilman 1997, Shurin 2000). Though assembly theory has primarily considered only how these interactions (i.e. predation and competition among species) result in biotic invasion resistance, others have indicated that these interactions can also explain patterns of coexistence (Case 1991, Brooks & Dodson 1965; Paine 1966; Connell 1983; Schoener 1983; Dobson & Hudson 1986; Kerfoot & Sih 1987). For
example, traits associated with competition for a shared resource may become
overdispersed and reduce niche overlap among species (Diamond 1975, Fox 1987, Fox and Brown 1993, Pacala and Tilman 1994, Morris and Knight 1996, Belyea and
Lancaster 1999). However, the extent to which biotic interactions will have a positive or negative influence on coexistence both within and among species has been debated. Some have suggested that when victim species (e.g. prey or host) share a common enemy (e.g. parasite or predator), competition for enemy free space may be intensified (Holt and Lawton 1994). Meanwhile, others have shown that enemies may relax competition, thereby enhancing coexistence among victims (Freeland 1983, Altermatt et al. 2007). However, the relative importance of these interspecific processes remains undetermined in most systems (Ricklefs and Schluter 1993, Hubbell 2001).
Generating accurate predictions regarding community assembly is further complicated by the fact that genetic variation in key traits (e.g., dispersal ability,
competitive ability, predation defense, etc.) within a species may also affect the outcome of inter-specific interactions and thus influence patterns of community assembly (De Meester et al. 2002, Weltzin et al. 2003, Stoll and Prati 2001, Hooper et al. 2005,
meta-population and variation in dispersal ability, spatially-variable selection may result in genetic differentiation among populations, based populations being founded from the clone/s that arrive first with the best fitness in a particular environment (Chesson 2000, De Meester et al. 2002, Holyoak et al. 2005). Furthermore, these priority effects have been extended to explain patterns of community-level species differentiation. For example, populations of early arriving colonists may exhibit rapid population growth thereby monopolizing community resources. This can decrease available niche space and consequently exclude later arriving genotypes and/or species (Vanormelingen et al. 2009, De Meester et al. 2002). With sufficient genetic variation within the meta-population, spatially-variable selection may then result in genotype sorting within a landscape (Chesson 2000, Holyoak et al. 2005). This form of priority effect may reduce local species richness and diversity has been proposed as a mechanism by which adjacent communities can exhibit both genetic and species differentiation (De Meester et al. 2002, Vanormelingen et al. 2009). Hence, variation in ecologically relevant traits may influence assembly dynamics and should receive future empirical attention (Bolnick et al. 2003, 2011, Vellend 2006, Urban and Skelly 2006, Urban 2011, Violle et al. 2012).
Pond metacommunities are a promising system to disentangle the relative
importance of colonization history and biotic interactions such as intra- and inter-species diversity on community assembly (Leibold and Norberg 2004, Holyoak et al. 2005, Urban and Skelly 2006, Urban et al. 2008, Loeuille and Leibold 2008, Urban and De Meester 2009). The zooplankton inhabiting temporary ponds are known for high rates of overland dispersal, yet there is often strong genetic differentiation among populations, suggesting that successful colonization may be minimal (Cáceres and Soluk 2002,
Cottenie et al. 2003, Cohen and Shurin 2003, Allen 2007, Vanschoenwinkel et al. 2008). This paradox was formalized by De Meester and colleagues (2002) in the
‘Monopolization Hypothesis’, which attempts to explain the simultaneously observed high rate of dispersal yet low gene flow among populations. This hypothesis predicts that rapid population growth by early arriving colonists (genotypes or species) can reduce subsequent colonization by invading individuals. Empirical examinations of this
hypothesis have been insufficient because they do not examine how this monopolization manifests itself through time (e.g., Thielsch et al. 2009) or are unreplicated and do not consider how the metacommunity scale may influence local patterns of multiple
developing communities (Ortells et al. 2014). Large scale field tests that simultaneously examine how both within and among species variation influence community
monopolization and how communities assemble over both time and space in newly established habitats are necessary.
We used a set of 38 experimental pools to test empirically if priority effects are generated by intra- and inter-species diversity of freshwater zooplankton and how this may influence community assembly. In some treatments, we attempted to control for priority effects induced by differences in dispersal abilities by relaxing dispersal limitation and stocking half of the ponds with species from the regional species pool. Genetic diversity of a focal grazer (Daphnia pulex) was crossed with this species diversity treatment by either adding one (no diversity) or six (high diversity) clones resulting in a 2x2 fully factorial and replicated design. The design also included 11 control pools with no species or D. pulex added. We hypothesized that stocking diversity would affect species richness and diversity during the three years of community assembly
examined in this study. Specifically, we predicted that 1) stocking ponds with regional species (high species diversity treatments) would relax dispersal limitation leading to A) increased species richness and diversity and B) communities reaching species saturation more rapidly than pools not stocked with the regional species pool (Tilman 1997, Shurin 2000) and 2) ‘high’ genetic diversity of a focal grazer species D. pulex would increase the likelihood of community monopolization thereby decreasing species richness and
diversity (De Meester et al. 2002, Thielsch et al. 2009) (see Figure 1.1). Given the
importance of dispersal in metacommunity theory, we also quantified the rate of overland dispersal (Cottenie and De Meester 2004, Leibold et al. 2004, Holyoak et al. 2005, Logue et al. 2011). We predicted that a greater number of source pools in a location (hexagon) would provide more dispersing propagules, thereby translating to differences in
community assembly dynamics (i.e. increased community similarity among pools in the same hexagon) compared to areas with fewer neighboring habitats, which would receive fewer dispersing propagules.
1.2Materials and Methods
Study System:
In summer 2010, through collaboration with the Upper Susquehana Coalition (USC), a network of 38 experimental pools were excavated at Svend O. Heiberg Memorial Forest (Tully, NY) and were filled naturally by rainfall. The pools were constructed in a 9.1ha forested plot and are a part of a larger 1578ha forest belonging to the State University of New York College of Environmental Science & Forestry (SUNY-ESF). Pools were designed to be 10m diameter circles and mimic natural vernal ponds.
Pools were constructed in one of three layouts (9, 3, or 1 pools per cluster) and delimited into arbitrary landscape “hexagons” which were 330m in width (See Appendix A.1 for spatial arrangement). The actual diameter and depths of pools when fully filled varied from 4.5m to 11m in diameter and 21cm to 67cm deep; some failed to dry during the study. Pools were not connected and did not overflow into adjacent pools at any point throughout this experiment. There were 29 pre-existing vernal ponds (both natural and anthropogenic) found near or within our experimental pool infrastructure (five within hexagon 20, eight within hexagon 9, two within hexagon 17, one in hexagon 15, and thirteen located 300m to 2km from the experimental pool infrastructure).
We used the crustacean Daphnia pulex as our focal species for manipulations of genetic diversity because it is a dominant species in many temporary ponds and exert considerable grazing pressure on algal resources (Brooks 1969, Shapiro et al. 1975). In an initial survey of 46 natural and anthropogenic ponds, 9 were found to contain Daphnia populations. We collected 61 D. pulex individuals from these 9 regional ponds and found that all individuals reproduced by obligate asexuality, each natural pond contained only 2-3 genotypes of D. pulex (D. pulex-pulicaria hybrids were also present in the region), and we found only 6 unique genotypes common in this region. We identified and isolated six unique D. pulex clones using six microsatellite marker loci for the Daphnia pulex (Dp) complex (Colbourne et al. 2004, Cristescu et al. 2006: Dp27, 78, 107, 196, 433, 461) which were used in the genetic diversity stocking treatments.
Experimental design:
To study the effects of initial stocking diversity on community assembly, on 18-21 May 2011, 27 experimental pools were stocked in a 2x2 fully factorial design (Appendix A.1) with two Daphnia pulex diversity treatments (no genetic diversity or ‘high’ genetic diversity) crossed with the addition of no other zooplankton species (no species diversity) or the addition of a regional zooplankton species inoculum consisting of 14 species after Daphnia were removed (high species diversity). Regional zooplankton were collected from 25 regional ponds, pooled, Daphnia removed, and 750 adult animals added to half (13) of the experimental pools (‘high’ species diversity) (see table 1.1 for taxa in initial inoculum and relative proportions added). Half of all experimental pools (13 pools) received no genetic diversity, which was represented by one D. pulex genotype, whereas, the other half (14 pools) were stocked with ‘high’ genetic diversity receiving either five or six unique genotypes of D. pulex (Chapter 2). All pools were stocked with 750 laboratory-reared D. pulex individuals (either 750 of one clone for no genetic diversity treatments, or 750 individuals divided among five or six focal clones).
Experimental pools were sampled for zooplankton twice per month (May to August) from May 2011 until the first sampling period in May 2013. We expected two years would be adequate to capture any temporal dynamics in aquatic zooplankton, as other studies have found community saturation on a similar timescale (Cáceres and Soluk 2002, Louette et al. 2008). To sample the zooplankton communities, three liters were taken from the center of each pool, filtered using an 80 μm sieve, and animals were preserved in 95% EtOH. All sampling gear was bleached between the sampling of each pool to ensure species were not accidentally transported among pools. Zooplankton were
counted and identified to the lowest taxonomic identity possible using “An Image-Based Key to the Zooplankton of the Northeast (USA)” (Versions 4.0 and 5.0: Hanley 2010, 2013). Each whole sample was scanned for rare taxa. Taxa with less than 300 individuals were counted completely, and those with over 300 individuals were subsampled (three, 2ml subsamples taken from 100ml diluted sample).
We used a 2-factor repeated measures ANOVA in SYSTAT 13 (SYSTAT Inc. 2010) to investigate the effects of time, genetic diversity, and species diversity on cumulative species richness (excluding the control pools). To avoid potential
confounding interactions between treatment and dispersal differences based on number of neighboring source pools, we analyzed both the entire data set, and only those pools in the 9 pool hexagons. In addition, we calculated the Shannon-Weaver diversity index for the final sampling date (May 2013) using the R’ Vegan Package (Oksanen et al. 2013) and used a two-way ANOVA in SYSTAT 13 to investigate the effects of initial species and genetic diversity on the May 2013 species diversity.
To determine if stocking diversity affected community structure after three years, community abundance and diversity data from May 2013 were imported into
Primer6/PERMANOVA + (Primer-E Ltd., Plymouth, United Kingdom) and square root transformed for normalization. A Bray-Curtis resemblance matrix was generated using community composition for the 38 pools in the experiment for multidimensional scaling (NMDS), PERMANOVA, and ANOSIM analyses. Bray-Curtis was employed because its measure is robust to the joint absences of a species in two samples, which is not the case for other coefficients (Clarke and Green 1988). We performed both PERMANOVA and ANOSIM analyses to determine if variation in community structure was explained by
location (the hexagon location for each pool). To explore the relationship between environmental variables (pH, temperature, dissolved oxygen concentration) and
zooplankton community structure, environmental variables for each pond were averaged across all sampling dates from 2011 to 2013. We then generated both a distance-based linear model (DistLM) and a distance-based redundancy analysis (dbRDA) in
Primer6/PERMANOVA+ to visualize results in multidimensional space.
To obtain estimates of overland dispersal (propagule pressure), in July 2012, four 26L containers (57cm x 41cm x 14cm) were placed flush with the ground in each of 6 hexagons (9 pool hexagons: 5 and 11, 3 pool hexagons: 9 and 10, and 1 pool hexagons: 8 and 14). Containers were located 5-20 meters from each neighboring pool, and were filled with ~15L reclaimed DI water (0.26 g/L, RO Right, Kent Marine, WI). Containers were destructively sampled (through an 80 μm sieve) after 7 days, and animals preserved in 95% EtOH. All animals and dormant eggs were counted and identified to the lowest taxonomic unit as above. Zooplankton propagules for each group (cladocerans, copepods, or all zooplankton) were summed across all 4 containers per hexagon and we used three one-way ANOVAs in SYSTAT 13 to examine the effect of number of nearest
neighboring source pools (9 vs 3 vs 1 neighboring pools) on 1) cladoceran propagules, 2) copepod propagules, and 3) all zooplankton propagules.
1.3 Results
The zooplankton assemblages added to the ‘high’ species diversity treatments were dominated by immature cyclopoid copepods, the majority of which most likely belonged to one of the three species of cyclopoid adult recorded (Table 1.1). Chydorus
sphericus was the dominant cladoceran. The initial inoculum contained at least 14 species of crustacean zooplankton; however, 7 species were never detected throughout the course of sampling. These non-persisting species were added at the lowest densities (<1% of initial inoculum), which may not have exceeded the minimum threshold of individuals needed to establish viable populations. Early successional communities were dominated by copepods in 2011, and by 2013 we observed a shift to more evenly distributed copepod and cladoceran communities, with D. pulex constituting 17.50% of sampled individuals in 2013 (Appendix A.2).
Despite the differing stocking treatments, all pools followed similar colonization curves with most species accumulating during the spring and summer months (Figure 1.2A). Even though a total of 31 taxa were recorded in the entire metacommunity by May 2013, the average cumulative taxonomic richness ranged from 6.1 (in a control pool) to a high of 7.6 (in a low genetic/low species diversity pool). The highest richness observed in any one pool was 13 taxa. Colonization rates were variable among pools and ranged from 0.98 to 4.9 colonists/year. The results of the rmANOVA revealed a significant effect of time, but no effect of species diversity, genetic diversity, or their interaction on
cumulative species richness (Figure 1.2, full statistics reported in Appendix B.1). Given the clustering of pools in the landscape, the analysis of all pools confounds diversity treatments with number of neighboring pools, hence we also analyzed the data only from those hexagons with 9 pools. A similar pattern emerged for the 9-pool hexagons
suggesting that the number of neighboring pools did not confound patterns of species accumulation (Figure 1.2B).
There was also no significant effect of species or genetic stocking diversity on the diversity index in May 2013 (Figure 1.3, full statistics reported in Appendix B.2). This result also held for the 9-pool hexagons. We found that variation in community structure in 2013 was not explained by initial stocking treatment, but was explained by location (hexagon) and select abiotic variables. Pools in the same hexagon tended to cluster to one another in multivariate space more closely than pools within the same treatment (Figure 1.4B and 1.4D, full statistics reported in Appendix B.3). Dissolved oxygen concentration (DO) and pH were variable among ponds and ranged from 0 to 30.09 mg/L, and 5.69 to 8.49, respectively and explained variation in zooplankton community structure in 2013, but average temperature did not (Appendix A.3). Clustering of pools in the by hexagon was less evident in the dbRDA (Appendix A.3), suggesting that zooplankton
communities in our system exhibited different species assemblages based on DO and pH, but not temperature which are only partially explained by differences among hexagons.
We found that dispersal occurred frequently across the experimental pool
landscape and recorded a total of 445 zooplankton propagules from 8 taxa in the dispersal traps (Copepods: Eucyclops agilis, Acanthocyclops spp., Tropocyclops prasinus
mexicanus, harparcticoid copepods; Cladocerans: Chydorus sphericus, Eurycercus spp, Ceriodaphnia ephippia, and Daphnia pulex individuals and ephippia), all of which colonized within 7 days. Some of the species found in our dispersal traps were not detected in the source pools during the time of this study (Eurycercus spp. and
Ceriodaphnia spp.). Samples from the source pools during this study reveal that locations (hexagons) with 9 neighboring source pools had more animals than locations with 3 or 1 neighboring pools (32,619, 13,242, and 2,074 total animals found during one sample
period in July 2013, respectively). This suggests that hexagons with 9 source ponds had a more potential propagules than 3 or 1 pool hexagons. However, the propagule pressure study revealed that the number of dispersing propagules (cladocerans only, copepods only, or all zooplankton) was not influenced by number of neighboring source pools (Figure 1.5, full statistics reported in Appendix B.4), which is likely explained by high variance between samples for both copepods and cladocerans. Number of cladocerans found in the traps ranged from 0 to 46, whereas copepods ranged from 6 to 225.
1.4 Discussion
We hypothesized that initial genetic and species diversity would affect assembly dynamics via priority effects. Specifically, we predicted that introducing species-rich communities into newly created pools would relax dispersal limitation, allowing these pools to have higher overall species richness and diversity while more rapidly reaching species saturation when compared to pools that were not stocked with the regional species. Additionally, we predicted that high genetic diversity of a focal grazer population, D. pulex, would result in the monopolization of algal resources, thereby decreasing species richness and diversity. However, by May 2013, we observed no significant differences in cumulative colonization rates, species richness or species diversity among treatment or control ponds. Community structure in 2013 did not differ based on treatment, but did differ among hexagons and was partially explained by pH and DO concentration. While dispersal was rapid, not all taxa were detected in traps possibly giving rise to hexagon-level differences in available species. As a result, community
differences may be explained by local abiotic factors or dispersal limitation as opposed to the diversity-related factors manipulated in this study.
Previous studies have provided mixed results regarding the role of history on community assembly (Diamond 1975, Neill 1975, Tilman et al. 1986, Drake 1991, Sommer 1991, Luh and Pimm 1993, Law and Morton 1996, King et al. 1996, Samuels and Drake 1997, Law 1999, Chase 2003). Some have shown that history primarily influences community assembly (Drake 1991, Inouye and Tilman 1995). However, the importance of local factors in shaping communities has been documented in multiple systems (Gliwicz and Pijanowska 1989, King et al. 1996, Vanormelingen et al. 2009). When local factors dominate community assembly, communities may converge on single state regardless of invasion order (Chase 2003). These processes need not be mutually exclusive and their relative importance is often context dependent (Chase 2010). Nevertheless, we suggest further large-scale field manipulations simultaneously addressing the relative importance of each of these mechanisms.
The relative proportion of taxonomic groups changed substantially from 2011 to 2013. We observed a shift from a copepod-dominated system to a more even distribution of copepods and cladocera. Previous studies have documented that cyclopoid copepods are among the first to colonize artificial ponds (Jenkins and Buikema 1998, Cáceres and Soluk 2002, Cohen and Shurin 2003, Frisch and Green 2007). Even though species richness increased annually, our system was less species rich (31 recorded zooplankton species) than other temporary pond systems (51 species: Mahoney et al. 1990, 53 species: Dodson and Silva-Briano 1996). The number of species in the regional pool can have a substantial effect on the predicted outcomes of community assembly (Chase 2003). Chase
(2003) summarized the evidence for multiple stable equilibria and provided a set of explicit predictions for when a single versus multiple community states should be observed. He predicted that local sites in less species -rich areas should converge on a single stable state. However, in our system, equilibrium may not yet be reached as our pool communities have not reached species saturation (see Figure 1.2). Other studies have demonstrated that community saturation in zooplankton can be achieved rapidly within 6 months to 2 years (Jenkins and Buikema 1998, Cáceres and Soluk 2002); Chase (2003) found in his zooplankton system that community composition was stable after 3 years of study. As a result, we cannot conclude that our communities are converging on a single community type, as our communities are still developing (i.e. not yet at a stable equilibrium). Further sampling will reveal whether our treatments promote multiple stable equilibria or convergence a single stable state.
Manipulations of species stocking diversity were intended to relax dispersal limitation, however, we did not see an effect of this manipulation on cumulative species richness, diversity, or community structure. Dispersal limitation has been shown on multiple occasions to reduce community richness and diversity (Cornell and Lawton 1992, Cornell and Karlson 1997, Shurin 2000). In the case of our study, all but five species stocked in the high diversity treatments failed to establish detectable populations throughout the entire course of this study. This may be explained by stochastic extinction resulting from many species in our inocula being present at very low proportions.
Furthermore, abiotic filters may have inhibited select species from establishing in early succession. Similar to our findings, Louette and De Meester (2005) found that out of 11 common regional species stocked to a set of newly formed ponds, 4 were absent or found
sporadically throughout the metacommunity. Previous studies have demonstrated that environmental factors determine establishment success of many taxa and play a large role in shaping communities (Gliwicz and Pijanowska 1989, King et al. 1996, Shurin 2000, Cottenie and De Meester 2004, Vanormelingen et al. 2009, Algar et al. 2011).
Nevertheless, the failure of these species to successfully colonize (e.g., “ghost species”) may have influenced the dynamics of community assembly and thus the final community structure (Miller et al. 2009).
We predicted that in the high-genetic diversity treatments there would be an increased likelihood that one or more introduced genotypes would exhibit high fitness in the pool, resulting in community monopolization and reduced species diversity over time (De Meester et al. 2002, Leibold et al. 2004, Urban et al. 2008, 2012, Urban and De Meester 2009). De Meester (2002) proposed that this monopolization may explain the paradox between rapid dispersal yet low gene flow which is observed in many freshwater zooplankton. In studies examining recently colonized aquatic habitats (< 5 years), there is evidence that founder events and priority effects in Daphnia lead to population genetic differentiation (Haag et al. 2006, Louette et al. 2007). Natural Daphnia metapopulations harbor extensive levels of genetic variation, which often translates to variation in
ecologically relevant life history traits (Lynch 1983, 1984a, Lynch et al. 1989). In an initial survey of the region, we attempted to capture all clonal variants for our genetic diversity treatments; however, in 2013 we did find a new clone in the experimental pools (Chapter 2). Despite our findings that populations were often dominated by one or a few genotypes (Chapter 2), we did not have evidence that stocking diversity influenced community diversity.
Given the importance of dispersal on metacommunity dynamics, we attempted to quantify dispersal occurring in our experimental infrastructure. Actual dispersal rates are difficult to quantify in passive and cryptic dispersers, however, many studies conducted on freshwater zooplankton have quantified dispersal mode, distance, and rate (Pajunen 1986, Holland and Jenkins 1998, Jenkins and Buikema 1998, Shurin 2000, Bilton et al. 2001, Cáceres and Soluk 2002, Cohen and Shurin 2003, Havel and Shurin 2004, Allen 2007). While we found reasonably high dispersal rates (eight taxa dispersing in 7 days), we found no difference in dispersal rates based on number of neighboring source pools. Of the 8 taxa observed in our traps, 2 taxa (Eurycercus spp. and Ceriodaphnia) went undetected in all sample ponds throughout the course of study, suggesting long distance dispersal events from ponds outside of our sample region or they were undetected in our pools during sampling. Zooplankton are capable of regular long distance dispersal events (20-100km; Shurin 2000). However, we cannot conclude that dispersal was not limiting in our communities, as only a small portion of the regional taxa were caught in our dispersal traps. Furthermore, pools tended to cluster in multivariate space by location, suggesting that dispersal may be rapidly occurring within the hexagon, but not over the entire experimental pool landscape.
Ecological theory has highlighted the importance of both intra-specific diversity and inter-specific diversity on community dynamics (Elton 1958, Tilman 1997, Shurin 2000, Loreau and Hector 2001, Fukami et al. 2005, Crutsinger et al. 2006, Bolnick et al. 2011, Violle et al. 2012). However, translating this theory to the field is difficult given the complex interaction of multiple factors operating at different hierarchical levels. Our study demonstrates that genetic diversity of a focal population and initial species
diversity does not appear to have an effect on early successional community species richness, diversity or community similarity in our planktonic metacommunity. Future large-scale manipulations simultaneously addressing multiple processes operating at multiple hierarchical levels (i.e. deterministic vs. stochastic, local vs. regional, intra- vs. inter- species variation) are required to better understand how communities assemble over both time and space.
1.5 Tables and Figures
Table 1.1: Illustrates the species composition and relative abundance during 3 time periods (2011, 2012, and May 2013) as well as the composition of the ‘high’ species diversity stocking aliquots. Numbers in parentheses are the number of taxa found in each category. Distribution shifted from a system dominated by cyclopoids in 2011, to a community more evenly distributed between copepod and cladocerans in 2013.
Regional Species Initially Stocked (12) Species found in 2011 (17) Species found in 2012 (21) Species found in May 2013 (13) Cladocerans Daphnia pulex (11.72%) Daphnia pulex (7.82%) Daphnia pulex (17.50%)
Chydorus sphericus (7.24%) Chydorus sphericus (0.35%) Chydorus sphericus (11.24%) Chydorus sphericus (25.13%)
Acroperus harpae (0.17%) Daphnia ambigua (0.03%) Sida crystallina (0.17%) Eurycercus spp. (0.21%) Bosmina longirostris (0.42%) Macrothricidae sp or Ilyocryptidae spp. (0.07%) Ceriodaphnia (0.03%) Daphnia dubia (<0.01%) Alona (0.01%)
Copepods Immature cyclopoids (69.67%) Immature cyclopoids (40.29%) Immature cyclopoids (33.79%) Immature cyclopoids (33.70%) Tropocyclops prasinus mexicanus (13.19%) Tropocyclops prasinus mexicanus (12.84%) Tropocyclops prasinus mexicanus (34.70%) Tropocyclops prasinus mexicanus (2.69%)
Microcyclops rubellus (3.17%) Microcyclops rubellus (8.98%) Microcyclops rubellus (0.27%) Microcyclops rubellus (0.03%) Eucyclops elegans (3.17%) Eucyclops elegans (0.37%) Eucyclops elegans (0.06%) Eucyclops elegans (0.20%)
Harpacticoid (1.40%) Harpacticoid (0.01%) Harpacticoid (<0.01%) Harpacticoid (0.15%) Skistodiaptomous oregonensis (<0.01%) Skistodiaptomous oregonensis (<0.01%) Eucyclops agilis (19.33%) Eucyclops agilis (6.01%) Eucyclops agilis (6.04%) Orthycyclps modestus (0.18%) Orthycyclps modestus (1.65%) Orthycyclps modestus (0.18%) Ectocyclops phaleratus (0.18%) Ectocyclops phaleratus (<0.01%)
Acanthocyclops spp. (0.92%) Acanthocyclops spp.(4.04%) Acanthocyclops spp. (10.41%) Unknown cyclopoid (13 antennal segment) (2.60%) Unknown cyclopoid (13 antennal segments) (0.07%)
Microcyclops vicarians (0.01%)
Immature calanoid (<0.01%)
Macrocyclops albidus (0.11%) Macrocyclops albidus (0.03%) Diacyclops bicuspidatus odessanus (0.12%)
Cyclops scotifer (0.01%)
Diacyclops thomasi (0.03%)
Macrocyclops fuscus (0.04%)
Figure 1.1: Predicted cumulative colonization curves for each stocking treatment (plus control). Rapid community saturation was predicted for pools stocked with regional species (high species diversity treatment). For those pools stocked with high Daphnia diversity, we predicted monopolization would decrease invasibility thereby resulting in less species rich communities.
Time (in weeks from stocking)
A
ve
ra
ge
C
umu
la
tive
T
axa
0
20
40
60
80
100
0
2
4
6
8
Time (weeks from inital stocking)
Ave
ra
g
e
C
u
mu
la
tive
T
a
xa
Low Daphnia Diversity, Low Species Diversity High Daphnia Diversity, Low Species Diversity Low Daphnia Diversity, High Species Diversity High Daphnia Diversity, High Species Diversity Control
0
20
40
60
80
100
0
2
4
6
8
Time (weeks from inital stocking)
Ave
ra
g
e
C
u
mu
la
tive
T
a
xa
B
!
A
!
Low#Daphnia#Diversity,#Low#Species#Diversity#
High#Daphnia#Diversity,#Low#Species#Diversity#
Low#Daphnia#Diversity,#High#Species#Diversity#
High#Daphnia#Diversity,#High#Species#Diversity#
Figure 1.2: Cumulative colonization curves for each stocking treatment (plus control) (+/- SE for final sampling date in May 2013). For each time period, the cumulative number of taxa was averaged across either: A. all ponds for each treatment in the experimental pond infrastructure or B. only across ponds in the 9-pool hexagons. Time 0 represents the week following stocking. 0 20 40 60 80 100 0 2 4 6 8
Time (weeks from inital stocking)
Ave ra g e C u mu la tive T a xa
Low Daphnia Diversity, Low Species Diversity
High Daphnia Diversity, Low Species Diversity
Low Daphnia Diversity, High Species Diversity High Daphnia Diversity, High Species Diversity Control 0 20 40 60 80 100 0 2 4 6 8
Time (weeks from inital stocking)
Ave ra g e C u mu la tive T a xa B! A! Low#Daphnia#Diversity,#Low#Species#Diversity# High#Daphnia#Diversity,#Low#Species#Diversity# Low#Daphnia#Diversity,#High#Species#Diversity# High#Daphnia#Diversity,#High#Species#Diversity# Control#
Figure 1.3: Shannon-Weaver diversity index calculated in May 2013 (+/- SE among ponds). A. All ponds in the experimental pool infrastructure; B. only ponds in the 9 pool hexagons. Black bars designate high species diversity treatments (stocked with 14 regional species) and white bars designate pools in which no regional species were stocked (low species diversity).
0! 0.2! 0.4! 0.6! 0.8! 1! None High Daphnia Diversity No Regional Species Regional Species Added
0 0.2 0.4 0.6 0.8 1 None High Sh an n o n -W ea ve r In d ex (H ’) Daphnia Diversity ! ! B! A!
Figure 1.4: NMDS plots of zooplankton communities for initial stocking (A and C) and field abundance in May 2013 (B and D). Panels A and C include labels by hexagon, whereas C and D include labels by treatment. Data for all graphs were square-root transformed. Relative distance on the NMDS plots reflect the relative dissimilarity between pool communities.
Ini@al#Stocking# May#2013# Hexagon# Treatment# A# B# C# D# J#Low#Species/Low#Daphnia#Diversity# J#Low#Species/High#Daphnia#Diversity# J#High#Species/Low#Daphnia#Diversity# J#High#Species/High#Daphnia#Diversity# J#Control # #######
Figure 1.5: Dispersal trap experiment after 7 days found a total of 445 propagules, but there was no difference in propagule abundance between major taxonomic groups (cladocera, copepods, or all taxa combined). Bars represent number of zooplankton propagules (cladocera [white bars], and copepods [black bars], and all zooplankton [black diagonal bars] averaged among replicate hexagons (+/- SE among replicates)).
0 50 100 150 200 250 1 3 9 N u m b er o f Pr o p ag u le s
Number of Neighboring Pools
Cladocerans Copepods All Zooplankton
CHAPTER 2: GENETIC DIFFERENTIATION OF DAPHNIA PULEX IN FRESHWATER PONDS: RAPID EVOLUTIONARY CHANGE AFTER TWO
YEARS 2.1 Introduction
Evolution on an ecological timescale (rapid evolution) is a widespread phenomenon in nature, and the reciprocal effects of rapid evolution and ecological processes (e.g., competition, predation, parasitism, etc.) can drive community dynamics over both time and space (Thompson 1998, Reznick and Ghalambor 2001, Hargeby et al. 2004, Gilchrist et al. 2004, Phillips et al. 2006, Carroll et al. 2007, Ghalambor et al. 2007, Schoener 2011, Ellner et al. 2011). For example, competition and predation not only influence community structure, but also can alter the evolutionary trajectory of
participating species (MacArthur 1972, May 1974, Fukami et al. 2007, Meyer and Kassen 2007, terHorst et al. 2010). Local adaptation of one or more participating species can then feed-back to alter the outcome of inter-specific interactions (De Meester et al. 2002, 2007, Steiner et al. 2007, Crutsinger et al. 2008, Bassar et al. 2010). These processes occurring in one local patch can affect the eco-evolutionary dynamics of other patches connected by dispersal, as predicted by the metacommunity framework (Leibold et al. 2004, Urban et al. 2008, 2012). However, field tests demonstrating the mechanisms by which genetic variation in a population influences the outcome of both intra- and inter- specific interactions during community assembly and the resulting spatial distribution of taxa remain rare in the literature.
In freshwater crustaceans, strong genetic differentiation is often observed among adjacent populations, despite frequent dispersal events and similar habitat types (Hebert
frequent dispersal, but limited successful colonization prompted De Meester and colleagues (2002) to develop the Monopolization hypothesis, a follow-up to Boileau et al’s (1992) persistent founder effects hypothesis. This model argues that local adaptation of founding individuals can lead to rapid population growth of one or a few genotypes. This evolution-mediated priority effect may result in monopolization of resources, thereby inhibiting later arriving colonists (De Meester et al. 2002). Theoretical and empirical tests of the monopolization hypothesis have shown that rapid local adaptation can indeed initiate priority effects, resulting in communities dominated by initial
colonists (De Meester et al. 2007, Urban and De Meester 2009, Ortells et al. 2014). Furthermore, Ortells et al. (2014) found that populations became dominated by early arriving genotypes, which then impeded invasion by other genotypes. However, we are not aware of any large-scale field studies that have examined if variation in initial genetic diversity can determine outcome of monopolization as hypothesized by De Meester et al. (2002).
Variation in phenotypic plasticity of ecologically relevant traits among genotypes may also influence the distribution of genotypes within a landscape (Thompson 1991, Lampert 1993, Tollrian 1995, Reboud and Bell 1997, Boersma et al. 1998, Kawecki and Ebert 2004, Allen et al. 2010). Certain genotypes (specialists) may do very well under a narrow range of environmental conditions but suffer severe fitness declines in other environments. On the other hand, fitness may be less variable for generalist genotypes, allowing them to invade more habitats. However, the ability of generalists to persist will depend on timing of arrival in relation to specialists; generalists may experience
specialists (Roughgarden 1972, Hedrick 1986, Smith and Hoekstra 2009). Specialists may be more variable in fitness across the landscape, and when in a non-suitable habitat will experience a fitness decline. However, when traits are a suitable match to the environment, these specialists will be competitively superior and displace generalists (Kassen 2002, Allen et al. 2010).
Obligate parthenogenic Daphnia are excellent model organisms with which to test the effects of initial genetic diversity on patterns of population monopolization. Without the opportunity to undergo recombination, as in sexually reproducing species, the genetic variation in asexual genotypes is “frozen” (Pantel et al. 2011a). As a result, evolution-mediated priority effects occur primarily through established trait variation rather than sexual reproduction producing novel combinations in separate local patches (De Meester 1996). Moreover, there is considerable phenotypic plasticity in morphological,
behavioral, and life-history traits, with some genotypes being more plastic than others (e.g., Weider and Pijanowska 1993, Boersma et al. 1998, Bertram et al. 2013). Given suitable traits, certain genotypes may successfully colonize one or more local patches and exhibit rapid population growth thereby excluding later arriving “invading” genotypes.
We established a field study to examine how initial genetic diversity of Daphnia pulex mayinfluence population persistence and landscape-level patterns of genetic structure after two years in newly created habitats. We stocked six genotypes, either alone or in combination, into 27 newly created freshwater pools. All populations were sampled every two weeks from May to August for the first two years and then once in the third year, at which time we determined the distribution and relative frequency of each genotype across the landscape. We used laboratory assays to estimate maximum intrinsic
rate of increase, hereafter clonal population growth rate (r), for each genotype on high –
quality food (the green alga Ankistrodesmus falcatus). To quantify plasticity in r in
response to different food levels, four of the six clones were assayed for r on both high (2 mg C / L) and low (1 mg C / L) food concentrations (Tessier and Consolatti 1991). We
predicted that genetically diverse D. pulex populations would be more likely to persist
after two years than non-genetically diverse populations and expected that patterns of local monopolization by particular genotypes in the field would correspond with
population growth rates and amount of plasticity measured in laboratory life-table assays.
2.2 Materials and Methods
To follow the landscape-level genetic change in a newly formed freshwater pool metapopulation, we experimentally manipulated initial genetic diversity of a focal
species, Daphnia pulex, in a set of 38 experimental pools created during the summer of
2010 at Svend O. Heiberg Memorial Forest (Tully, NY). The experimental pools were designed to be 10m diameter circles and were developed in a 9.1 ha forested plot belonging to the State University of New York College of Environmental Science & Forestry (SUNY-ESF). However, actual diameters when fully filled varied from 4.5m to 11m. Pools were not physically connected at any point throughout the course of this
experiment, but we found rapid dispersal of D. pulex in this landscape (Chapter 1). In
addition, there were 29 pre-existing vernal ponds found near or within the experimental pond infrastructure that may have served as sources of dispersing propagules (five within hexagon 20, eight within hexagon 9, two within hexagon 17, one in hexagon 15, and thirteen located >300 meters from the experimental pool infrastructure).
We used the crustacean Daphnia pulex as our focal species for manipulations of genetic diversity because they are a keystone species in many freshwater aquatic systems (Lampert 1987, Carpenter et al. 1987, Persson et al. 2007), have a sequenced genome (Colbourne et al. 2011), and numerous microsatellite loci have been developed for population genetics studies (Colbourne et al. 2004, Cristescu et al. 2006). An initial
survey of 61 D. pulex individuals from 9 nearby natural ponds containing Daphnia
populations documented relatively low amounts of genetic diversity, only 2-3 unique genotypes per pond (distinguished at 6 microsatellite loci), and obligate asexuality of all genotypes. In May 2011 (May 18-21), 27 out of the 38 experimental pools were stocked
with one of two D. pulex diversity treatments (no genetic diversity or ‘high’ genetic
diversity). The no genetic diversity treatment was represented by one clone (each pool receiving one randomly chosen clone), since that is the minimum number of genotypes that can found a population, whereas, the ‘high’ genetic diversity was represented by 6 genotypes collected from the regional pool. We originally believed we had 7 unique genotypes, however an error in genotyping was discovered after the pools were stocked, reducing the number of focal clones from 7 to 6. Hence, some pools had twice as many individuals for one genotype. To address questions of another study, half of the pools for each treatment also received a regional zooplankton species inoculum consisting of 14
species after Daphnia were removed. The remaining 12 pools received no stocked
individuals and served as a control.
Experimental pools were sampled every two weeks for zooplankton from May to
August in 2011 and 2012, and once in May 2013, and preserved in 95% EtOH. Daphnia
sampling. From those persisting populations, we genotyped (details below) up to thirty D. pulex individuals per pool in May 2013 to document change in genetic structure since stocking in 2011. A chi-square goodness of fit test was used to determine if the relative proportion of each genotype in the total metapopulation had changed after two years. To determine if clonal diversity of our focal population or species diversity influenced D. pulex population persistence after two years, we used a logistic ANOVA with an integrated chi-square goodness of fit test in SAS version 9.3.
Laboratory Assays:
To determine how genotypes differed in ecologically relevant traits, we performed a life-table assay on the genotypes stocked in the field experiment. Two replicate sublines for each genotype were reared for 3 generations under constant conditions: 20°C, 14 h day length, and fed high – quality (Ankistrodesmus falcatus) and high quantity (2 mg C / L) algae to standardize maternal effects (Tessier and Consolatti 1991). Eight female progeny (<16 hours old) were collected from each subline and raised under the same conditions described above, observed and fed daily. For all individuals, we recorded the age at maturity, age at releasing the first 5 clutches, and size of the first 5 clutches. Using these data we calculated the clonal population growth rate (r) for each individual (Tessier and Consolatti 1991). We used a nested ANOVA in SYSTAT 13 to examine how r differed among genotypes. Subline was treated as a random effect and was nested within genotype, our main fixed factor. We used a Tukey's HSD post hoc test to examine the pairwise comparisons between genotypes. To determine the extent to
which phenotypic variation was explained by genetic effects, the broad-sense heritability (H2
) was calculated using methods from Lynch and Walsh (1998).
A second life-table assay was conducted to determine if clones exhibited
plasticity in clonal growth rate in response to differing food quantity. Prior to the assay, maternal effects were standardized as described in the previous assay. Following
standardization, up to 10 individuals from each genotype were reared in one of two food quantity treatment groups (both high - quality Ankistrodesmus falcatus); individuals in the high food treatment were fed 2 mg Carbon /L every other day, and those in the low food treatment were fed 1 mg C /L every other day (Tessier and Consolatti 1991). Experimental animals were incubated at 20°C with 14 h day length and we recorded the age at which each animal reached maturity, the age at releasing the first 4 clutches, and size of the first 4 clutches which were used to calculate r (Tessier and Consolatti 1991). We conducted a two-way ANOVA in SYSTAT 13 to examine the effect of genotype, food quantity, and the interactions (genotype × food quantity) on clonal population growth rate. A Tukey’s HSD post-hoc test was run using the interactive effect as the main model to examine the pairwise comparisons between high and low food quantity for each clonal genotype. Clones were considered to exhibit plasticity for r if they exhibited significant differences in growth rates between high and low food quantity treatments.
Molecular methods:
We used microsatellite markersto select focal D. pulex genotypes and to examine how the genetic structure of each D. pulex population changed after two years in our experimental pool infrastructure. Tissue digestion and DNA extraction were completed
using a Qiagen DNeasy® Blood & Tissue Kit. For tissue digestion, 20µl of proteinase K and 180µl of Buffer ATL (tissue lysis buffer) were added to each tube containing an individual D. pulex and incubated at 37°C for 12 to 20 hours. Digested tissue was
transferred to a spin column containing 200µl EtOH and 200µl AL solution (lysis buffer). The mixture was centrifuged at 8000rpm for one minute and the filtrate washed using Buffers AW1 and AW2. The DNA was collected in a clean tube with 30µl Buffer AE (elution buffer). Six loci for the Daphnia pulex (Dp) complex (Dp27, 78, 107, 196, 433, 461: Colbourne et al. 2004, Cristescu et al. 2006) were amplified using Qiagen a
multiplex PCR kit. 1µl of extracted DNA was added to 6µl Qiagen multiplex PCR
mastermix, 1.2µl of Dp primer mix, and 3.8µl of sterile molecular grade water and run on a DNA Engine Dyad® Thermal Cycler. Cycling conditions were initiated with one cycle at 95 °C for 15 min, followed by 30 cycles of (94 °C for 30 s, 50.5 °C for 180 s, 72 °C for 90 s) and a final extension at 72 °C for 10 min. Amplified DNA was diluted (1µl
amplified DNA and 10µl sterile molecular grade water) and sent to the W.M. Keck Center for Comparative and Functional Genomics at the University of Illinois at Urbana-Champaign Biotechnology Center (Urbana, IL, USA) for a microsatellite fragment analysis. Fragment analysis was conducted using GeneMapper™ Version 3.7 software (Applied Biosystems, Foster City, CA, USA).
2.3 Results
The first Daphnia were recorded one week after stocking in 15 out of the 39 pools (across both treatments and control), yet by 2013, 4 pools failed to establish Daphnia populations (detectable Daphnia pulex found in samples) at any point throughout the
study (three in hexagon 20, and one in hexagon 10). Of the 14 pools stocked with high
Daphnia diversity, 10 populations persisted, whereas nine populations (of 13 stocked) persisted in pools stocked with no Daphnia diversity. As a result, we found no difference in Daphnia population persistence between no or high genetic diversity treatments (Table 2.1, N = 27, χ² = 0.03, p = 0.86) and no effect of stocking ponds with regional species on
Daphnia persistence (N = 27, χ² = 0.96, p = 0.33).
Genotypic frequencies in the metapopulation changed significantly from stocking to May 2013 (Figure 2.1, N = 6, χ² = 716.7, p < 0.0001). Clones 1, 2, and 3 dominated the landscape in May 2013, often in pools in which they were not originally stocked. Clone 1 was most abundant in the landscape and was found in 23 out of 27 ponds with
Daphnia populations (Figure 2.1). Clone 3 was found almost exclusively in 3 adjacent hexagons at the south east corner of the forest (hexagons 5, 9, and 11). Two genotypes (5 and 6) were not detected in May 2013 and an unidentified genotype was found in our landscape (termed ‘New Clone’ found in hexagon 9: Figure 2.1). Persistence of
individual genotypes was not influenced by whether they were stocked alone or in a more genetically diverse community (Table 2.1, N = 76, χ² = 0.010, p = 0.911). We had 30 ‘invasions’ in which a genotype was found in a pond in which it was not originally stocked, including one new clone that was not originally stocked.
When raised in the laboratory under high food quantity and quality, the genotypes used in the field experiment exhibited small but significant differences in maximal clonal growth rate (Figure 2.2, F5,6 = 7.59 p = 0.014). A Tukey’s HSD post-hoc test determined that clone 3 had a higher growth rate than clones 1, 2, 4, and 5 and this genotype was the second most abundant genotype in the field. However, the growth rate of the most
dominant genotype in the field (clone 1) was not significantly different from the 3 worst performing genotypes in the field (clones 4, 5, and 6). Furthermore, there was no
correlation between dominance in the field and clonal population growth rate measured in laboratory assays (R2
= 0.077, N = 4, p = 0.59). Much of the phenotypic variation in maximal clonal growth rate was attributed to genetic differences as broadsense heritability was very high (H2 = 0.718). However, this difference was not driven by differences in age or size at maturity (Appendix C.1).
Two of the clones (clones 1 and 5) failed to participate in the second laboratory assay assessing plasticity for r among clones. Of those clones that participated (clones 2, 3, 4, and 6), r ranged from 0.31 to 0.34 (individuals/individual * day) on high food quantity, and 0.17 to 0.25 (individuals/individual*day) on low food quantity.
Measurements for r collected on the high food quantity in this assay, correspond with well with r measurements in the first assay. We found a significant effect of clonal genotype, food quantity, and the G×E interaction on clonal growth rate (full statistics reported in Appendix C.2). Three clones (2, 3 and 4) exhibited plasticity in clonal growth rate on varying food quantity (Appendix C.2). For each of these three clones, growth rate was higher when raised on high food than on low food; however, clone 6 was unaffected by low food.
2.4 Discussion
We demonstrate that clonal selection in an asexual metapopulation of Daphnia pulex has resulted in rapid changes in population genetic structure in newly established freshwater ponds. We predicted that genetically diverse D. pulex populations would be
more likely to persist when compared to non-genetically diverse populations and performance in the field (ability to persist) would correspond with performance (r) in laboratory assays. We did not find that genetically diverse populations were more likely to persist than non-genetically diverse populations. Nevertheless, the distribution of genotypes in the landscape changed significantly over the two years of this study. Many pools came to be dominated by one or a few genotypes and in multiple instances, these were genotypes not originally stocked. In addition, we detected a new clone in our experimental pool infrastructure that likely dispersed in from a nearby natural pond. Results from life-table assays suggest these observed changes in the field do not correspond with laboratory estimates of growth rates on high-quality food, and dominance in the field was not correlated with maximal growth rate. Changes in
population genetic structure of D. pulex is possibly driven by a few specialist genotypes that, while they showed sensitivity to different resource quantities in lab, exhibited traits which favored their establishment in the experimental pools.
Multiple studies have shown that local biotic and abiotic factors can influence clonal distribution in freshwater cladocerans (Lynch 1983, Weider 1985, Carvalho and Crisp 1987, Weider and Hebert 1987, Pantel et al. 2011b), and several pieces of evidence suggest that local factors may be responsible for the current patterns that we observe. First, we found that dispersal is rapid in this experimental pool landscape (0-46 D. pulex dispersing in 7 days: Chapter 1), suggesting that at least within each hexagon, dispersal does not appear to be liming. The fact that there were 30 “invasions” where a clone that was not originally stocked in a pool was found in that pool in 2013, is further indication of ongoing dispersal. Of the 78 stocking events (six genotypes into 27 ponds), there were
56 times in which a genotype originally stocked was not detected in May 2013, indicating either local extinction or a population size below detection limit during May 2013.
However, despite rapid dispersal, our metapopulation was dominated by three clonal genotypes in May 2013 suggesting colonization ability into each pond varied among genotypes and may be partially explained by local abiotic filtering. This has been
observed in a fishless lake population of Daphnia magna, where changes in clonal
frequencies followed seasonal environment changes (Carvalho and Crisp 1987). In a previous study, we found that among-pond variation in community structure was partially explained by pH and dissolved oxygen concentration (DO); these factors may also
determine the distribution of D. pulex populations in this study (Chapter 1). As a result,
spatially-variable clonal selection may drive the distribution of genotypes in our
landscape favoring the persistence of genotypes exhibiting traits suitable for a particular pool.
We expected that genetic diversity would enhance population monopolization as described in De Meester and colleagues’ Monopolization Hypothesis (2002). While others have found significant evidence for a strong founder effect contributing to population monopolization (Louette et al. 2007, Ortells et al. 2014), the lack of
correlation between laboratory assays and patterns of persistence in the field give us no evidence that population monopolization resulted from an evolution-mediated priority effect. We did find that our populations became dominated by one or a few genotypes; however, this was not due to genotypic differences in r calculated in the laboratory assays. We expected a high clonal population growth rate would result in rapid
that heritability for r was higher (0.718) in our laboratory assays when compared to other studies documenting H2
of ecologically relevant traits in Daphnia we thought this trait
would translate well to performance in the field pools. For example, average heritability ranges from 0.30 - 0.55 for clutch size (Lynch 1987, Lynch et al. 1989), 0.37 - 0.42 and 0.31 - 0.60 for body size in pond and lake populations, respectively (Lynch 1984b, Leibold and Tessier 1991, Tessier et al. 1992, Spitze 1993), and 0.34 - 0.35 for adult growth rate (Lynch 1984b, Spitze 1993). Clonal differences in r were highly heritable in relation to other ecologically relevant traits measured in Daphnia, but conditions in the
laboratory assay may differ from conditions experienced in the ponds, explaining why laboratory performance did not correspond with field performance. As a result, we conclude that monopolization was not likely driven by an evolution-mediated priority effect based on differences in clonal growth rates, but rather was possibly driven by the local abiotic filtering of genotypes possibly based on varying levels of phenotypic plasticity.
Daphnia show plasticity in multiple life history traits and inter-clonal lineages
may vary in their degrees of plasticity (Tessier and Consolatti 1991). As a result, inter-clonal lineages may differ in their ability to tolerate environmental conditions. Allen et al. (2010) found that plasticity in ecologically relevant traits, as opposed to local adaptation, explained patterns genotype distribution of D. pulex in freshwater ponds. Our preliminary
results demonstrating among-clone variation in plasticity suggest a potential explanation for the patterns of colonization observed in this study. Three of the persisting clones in the field measured in our second life-table assay (clones 2, 3, and 4), exhibited plasticity in r in response to food quantity, with high growth rates observed in the presence of high
food, but much reduced growth rates in low food. We found one genotype that did equally well in both laboratory environments, but was not observed in field samples during May 2013. While others have suggested that generalist genotypes can rapidly colonize new environments, we are unable to determine if the clones present in May 2013 represent the composition of the “early” colonists or are actually specialists that displaced early colonizers (Reboud and Bell 1997, Kassen 2002, Allen et al. 2010). Given the results of the plasticity assay, we give more support for the latter explanation. However, the temporal dynamics of genotype assemblage is unclear at this point in time given that we only observed genotype composition for one sampling date. Future field and
laboratory measurements will provide increased resolution regarding possible
environmental factors and the role of plasticity within a species has in determining the distribution of particular genotypes in the field over both time and space.
Initial genetic stocking diversity of D. pulex populations did not affect population persistence; however, populations exhibited rapid evolutionary change after only two years. The experimental landscape became dominated by three clones (1, 2, and 3), while two of the originally stocked clones vanished from the landscape (5 and 6). Laboratory assays on ecologically relevant traits results provided an incomplete explanation of observed patterns in the field. We conclude that performance in the field is possibly explained by local abiotic filtering and/or variation in plasticity among clones (i.e. generalist versus specialist genotypes). While we certainly do not argue that evolution-mediated priority effects do not occur in freshwater zooplankton systems, they did not appear to underlie the patterns of monopolization occurred in our system. Future studies on the mechanisms underlying landscape genetic structure should explicitly take into
account the role that phenotypic plasticity and various abiotic factors have in explaining patterns of genotype distribution across a landscape.
2.5 Tables and Figures
Logistic ANOVA : χ² Goodness of Fit
Source N df χ² P
Species Diversity Treatment 76 1 0.010 0.911
Genetic Diversity Treatment 78 1 1.680 0.195
Table 2.1: The number of ponds each genotype was stocked into by stocking treatment (either no or high D. pulex diversity) and the number in which that genotype went extinct by 2013. We also show the number of times a genotype invaded a new pool.!Neither! genetic nor species stocking diversity affected genotype persistence.
! No D. pulex Diversity! 'High' D. pulex Diversity! Invasions to Non-stocked Pools! Genotype! Stocked! Went Extinct! Stocked! Went Extinct! !
1! 2! 0! 10! 4! 13! 2! 5! 3! 14! 7! 7! 3! 1! 1! 10! 7! 9! 4! 2! 0! 10! 10! 0! 5! 2! 2! 10! 10! 0! 6! 2! 2! 10! 10! 0! New Clone! 0! 0! 0! 0! 1!
Figure 2.1: Genotype frequencies in experimental pool landscape of A. May 2011 genetic diversity stocking inoculums and B. two years post-stocking in May 2013. Each pool is represented by a small pie chart and the genetic structure for the entire metapopulation is represented by the large pie chart to the left of the experimental pool infrastructure.
Metacommunity Metacommunity
A
B
Clone: 1 2 3 4 5 6 New CloneFigure 2.2: Genotype-level differences for the calculated clonal population growth rate (r). Bars represent SE among sublines. Different lower-case letters on bars represent statistically significant differences (p<0.05) after ANOVA and Tukey’s HSD post hoc test.
